A&A 407, 31-40 (2003)
DOI: 10.1051/0004-6361:20030826
D. Proust1 - H. V. Capelato2 - G. Hickel2,3 - L. Sodré Jr.4 - G. B. Lima Neto4 - H. Cuevas4,5
1 -
Observatoire de Paris-Meudon, GEPI, 92195 Meudon, France
2 - Divisão de Astrofísica, INPE/MCT, 12227-010,
São José dos Campos/SP, Brazil
3 - Universidad do Vale do Paraoba, Av. S. Hifumi 2911,
São José dos Campos/SP, Brazil
4 - Instituto de Astronomia, Geofísica e Ciências Atmos.,
Universidade de São Paulo (IAG/USP), 01065-970 São Paulo/SP, Brazil
5 - Universidad La Serena, Casilla 544, La Serena, Chile
Received 18 April 2003 / Accepted 22 May 2003
Abstract
We present a dynamical analysis of the galaxy cluster Abell 376 based on a set
of 73 velocities, most of them measured at Pic du Midi and Haute-Provence
observatories and completed with data from the literature. Data on individual
galaxies are presented and the accuracy of the determined velocities is
discussed as well as some properties of the cluster. We obtained an improved
mean redshift value
z = 0.0478+0.005-0.006 and velocity dispersion
.
Our analysis indicates that inside a radius of
900 h70-1 kpc
(
15 arcmin) the cluster is well relaxed without any remarkable features
and the X-ray emission traces fairly well the galaxy distribution. A possible
substructure is seen at 20 arcmin from the centre towards the Southwest
direction, but is not confirmed by the velocity field. This SW clump is,
however, kinematically bound to the main structure of Abell 376. A dense
condensation of galaxies is detected at 46 arcmin (projected distance
2.6
h70-1 Mpc) from the centre towards the Northwest and analysis of the
apparent luminosity distribution of its galaxies suggests that this clump is
part of the large scale structure of Abell 376.
X-ray spectroscopic analysis of ASCA data resulted in a temperature
keV and metal abundance
.
The velocity
dispersion corresponding to this temperature using the TX-
scaling relation is in agreement with the measured galaxies velocities.
Key words: galaxies: distances and redshifts - galaxies: clusters: general - galaxies: clusters: individual: Abell 376 - X-rays: galaxies: clusters
Redshift surveys in clusters of galaxies are needed to study their dynamical and evolutionary states, estimating parameters such as the mass, shape and distortion of the velocity field, presence of substructures or projected foreground/background galaxies and groups, strength of dynamical friction and two-body processes and, in general, the present stage of their dynamical evolution. This information is useful to test theoretical scenarios of galaxy formation and also of large structure formation and evolution.
In clusters, the mean velocity is a key factor in deriving distances, allowing the study of matter distribution over very large scales. Within clusters, the analysis of the velocity field can lead to an estimate of the virial mass, constraining models of the dark matter content. Galaxy velocity measurements provide information complementary to that obtained through X-ray observations of clusters. Both form basic pieces of information for the understanding of clusters. However, a discrepancy between these estimators often found (e.g. Mushotzky et al. 1995; Girardi et al. 1998; Wu et al. 1999). Virial mass estimates rely on the assumption of dynamical equilibrium. X-ray mass estimates also depend on the dynamical equilibrium hypothesis and on the still not well-constrained intra-cluster gas temperature gradient (e.g., De Grandi et al. 2002). Finally, mass estimates based on gravitational strong lensing are considered more reliable than the others (e.g., Mellier 1999) because they are completely independent of the dynamical status of the cluster. The drawback is that strong lensing can only probe the central region of clusters. The discrepancies with other methods may be due to non-equilibrium effects in the central region of the clusters (Allen 1998).
An important source of departure from equilibrium (that may affect mass estimates) are substructures. Their very existence supports the current view that clusters grow hierarchically by accreting nearby groups and galaxies. Even the frequency and degree of clumpiness in the central regions of the clusters depends on the cosmology (e.g. Richstone et al. 1992). In many cases, substructures are loosely bound and can survive only a few crossing times in the hostile environment of rich clusters. However, they seem to be very common in present-day clusters. A recent estimate by Kolokotronis et al. (2000) indicates that at least 45% of rich clusters present optical substructures. Substructures are detected in both optical and X-ray images in 23% of the clusters. This last number may then be considered a lower limit on the fraction of real substructures in clusters and implies that one in four clusters may be out of equilibrium due to the presence of a substructure. The dynamical status of individual clusters should therefore be examined in detail before being used in other studies.
In this paper, we present a study of the dynamical status of the cluster
Abell 376, from the analysis of the positions and velocities of its galaxies,
as well as from the intra-cluster gas X-ray emission. Abell 376 has a I-II
morphology in the Bautz-Morgan classification and a richness class R=0extending on 151 arcmin in diameter (Abell et al. 1989). It is a cD or D-galaxy
dominated cluster; its brightest cluster member (BCG) was recently imaged by
the HTS WFPC2 (Laine et al. 2003). It shows a smooth brightness profile with a
central core, although it is not well fitted by the so-called Nuker-law (Lauer
et al. 1995). Abell 376 was observed in X-rays by several satellites. It has a
moderate cooling flow
yr-1 (White
et al. 1997).
A search in the NED database indicates that only few radial velocities were known in the
field of the cluster. A list of 13 velocities is published in Capelato et al.
(1991) obtained at Haute-Provence observatory with the multiobject
spectrograph SILFID for which we noticed a few years after that except the
central E/D number 71 galaxy, there was misidentification in the
correspondence object/fiber/spectrum, so that the velocities cannot be used
and should be removed from any database. For this reason, we decided to
reobserve some of the previous galaxies and not take into account our previous
velocities. Here we examine some properties of the cluster Abell 376, using a
set of 73 new and already published velocities. The observations of radial
velocities reported here are part of a program to study the dynamical
structure of clusters of galaxies, started some years ago and with several
results already published (see e.g. Proust et al. 1992, 1995, 2000; Capelato
et al. 1991; Sodré et al. 2001).
This paper is organized as follows. We present in Sect. 2 the details of the
observations and data reduction. In Sects. 3 and 4 we discuss the distribution
and the velocity analysis of the cluster galaxies respectively. In Sect. 5 we
analyze the X-ray emission and gas distribution and in Sect. 6 the spectral
and morphological classification of the cluster galaxies. In Sect. 7 we
present mass estimates for the central region of the cluster, derived from
optical and X-ray observations and the dynamical status of Abell 376. Finally,
in Sect. 8 we summarize our conclusions. We adopt here, whenever necessary,
Mpc-1,
and
The new velocities presented in this paper have been obtained with the 1.93-m telescope at Haute-Provence Observatory and with the 2.0-m telescope at Pic du Midi (France).
Observations with the 1.93-m Haute-Provence Observatory telescope were carried
out in November 1997 and 1998. We used the CARELEC spectrograph at the
Cassegrain focus, equipped with a 150 lines/mm grating blazed at 5000 Å
and coupled to an EEV CCD detector
pixels with pixel size of
13.5
m. A dispersion of 260 Å/mm was used, providing spectral
coverage from 3600 to 7300 Å. Wavelength calibration was done using
exposures of He-Ne lamps.
We completed the observations during an observing run at the 2.0-m Bernard Lyot
telescope at Pic du Midi observatory in January 1997 with the ISARD
spectrograph in its long-slit mode with a dispersion of 233 Å/mm with the
TEK chip of
pixels of 25
m corresponding to 5.8 Å/pixel.
Typically two exposures of 2700 s each were taken for fields across the
cluster. Wavelength calibration was done using Hg-Ne lamps before and after
each exposure.
The data reduction was carried out with IRAF using the LONG SLIT package.
The spectra were rebinned with a scale of 1 Å/bin equally spaced in log
wavelength. Radial velocities have been determined using the cross-correlation
technique (Tonry & Davis 1979) implemented in the RVSAO package (Kurtz et al.
1991; Mink et al. 1995) with radial velocity standards obtained from
observations of late-type stars and previously well-studied galaxies.
A total of 71 velocities has been obtained from our observations.
Table 1 lists positions and heliocentric velocities for 73 individual galaxies in the cluster including data from Wegner et al. (1999) and
from Postman & Lauer (1995), with the following columns:
Table 1: Heliocentric redshifts, position and morphological type for galaxies of Abell 376.
We present in Fig. 1 the adaptative density map of the projected
distribution of galaxies brighter than
obtained from the POSS I
Revised APS Catalogue
, in a
arcmin2 region centered on the position of the brightest cluster
member of Abell 376, E/D galaxy number 71 of Table 1. This corresponds to a
region of about
Mpc2 at the distance of
Abell 376 (z = 0.048, see below). The central square region, of about
arcmin2, indicates the area where redshifts have been measured
(see previous section). The positions of the brightest galaxies have been
superimposed to the density map.
![]() |
Figure 1:
Projected density map of the 168 galaxies brighter
than
![]() |
Open with DEXTER |
As it can be seen from this figure, the entire field is filled with structures among which Abell 376 stands out as the central dominant one. The density map suggests that Abell 376 has a regular galaxy distribution, flattened in the NE-SW direction. A substructure is present at SW, which is centered on a clump of bright galaxies around galaxy number 28, classified as E by Dressler (1980). As discussed in the next section, this clump may be considered as kinematically belonging to Abell 376 cluster. This figure also indicates that the cluster radial extension may attain several Mpc.
However, since Abell 376 lies close to the galactic plane (
), before trying to discuss its apparent projected galaxy
distribution, we should be confident that the galactic extinction is not
significantly affecting the magnitudes and counts. Figure 2 shows the map of
galactic absorption estimated from Schlegel et al. (1998), with a resolution of
2.5 arcmin. The colour excess E(B-V) was transformed to the O-magnitudes
using the appropriate transmission curve from Larsen (1996), given in the APS
site
and the
Table 6 from Schlegel et al. (1998). As it can be seen, although this map bears
a vague resemblance with density the distribution of galaxies given in
Fig. 1, the absorption AO never attains values higher than
0.4 and
thus it should not significantly affect the apparent distribution of galaxies.
![]() |
Figure 2:
Contours of constant galactic absorption, AO (labelled),
superimposed to the surface density image of galaxies brighter
than
![]() |
Open with DEXTER |
![]() |
Figure 3: O magnitudes histograms: continuous line: galaxies inside the square region of measured redshifts; the dashed line histogram is for galaxies with measured redshifts (see Fig. 1). Dotted lines: galaxies belonging to the NW condensation. |
Open with DEXTER |
![]() |
Figure 4:
Isopleths of the projected distribution of galaxies with
![]() ![]() |
Open with DEXTER |
Considering only galaxies brighter than
,
the central peak of
density is displaced by
4.7 arcmin E, relative to the central E/D
galaxy. However there is no such a displacement when considering galaxies
brighter than
(see also Fig. 4). For
,
central
densities amount to
Mpc-2).
There is no redshift information about the dense NW condensation seen in Fig. 1 since redshifts measurements are restricted in a square region around Abell 376. However a hint about its relative distance to Abell 376 may be given by comparing the apparent magnitude distribution of galaxies in Abell 376 with that of the NW condensation. This comparison is made in Fig. 3, where we also show the magnitude distribution of galaxies with measured redshifts. As it can be seen, both histograms, for Abell 376 and for the NW condensation, show almost the same slope, suggesting that both are about the same distance and probably belong to the same superstructure.
The histogram in Fig. 3 also indicates that the sample of measured radial
velocities in the central region of Abell 376 is 60% complete at
.
This is indeed a very shallow magnitude limit, given the
distance of this cluster. Figure 4 shows a comparison of the surface density
map of projected galaxies brighter than
together with the redshift
sample of galaxies discussed above, which, as discussed in the next Section,
comprises galaxies kinematically members of the cluster. As it can be seen,
excepted in the SE direction, the galaxies with measured redshift sample well
the projected galaxy distribution. Therefore, despite the incompleteness of
the redshift sample, it will be possible to discuss the line of sight (l.o.s.)
velocity distribution at least in a qualitative way.
Including previous measurements (Capelato et al. 1991; Postman & Lauer
1995; Wegner et al. 1999), there is a total of 73 measured velocities for
Abell 376. Using this sample we will discuss the cluster velocity distribution,
looking for non-equilibrium effects. The wedge diagrams of galaxies in right
ascension and declination are displayed in Fig. 5. Both show the cluster
and all galaxies collected from the NED database in a radius of 300 arcmin
around the cluster centre (476 objects). One can see that Abell 376 does not show
any connection or extension with other evident structure.
![]() |
Figure 5:
Wedge velocity diagram in right ascension (top), and
declination (bottom) for the galaxies in the Abell 376 field with
measured radial velocities smaller than
![]() ![]() |
Open with DEXTER |
We used the ROSTAT package (Beers et al. 1990; Bird & Beers 1993) to analyze
the distribution of the 63 galaxies in the selected velocity range
(for galaxies with more than one value of radial velocity in
Table 1, a mean value was assumed). In order to quantify the central location
and scale of the velocity distribution for Abell 376, we have used the
resistant and robust bi-weight estimators (
and
,
respectively) recommended by Beers et al. (1990). For this sample of velocities,
we obtain
and
.
Estimates of these quantities obtained with alternative
estimators show similar values. The one-sigma errors in these quantities are
calculated in ROSTAT by bootstrap re-sampling of 1000 subsamples of the
velocity data. For comparison, the radial velocity of the E/D galaxy located
at the centre of the main cluster is
,
near that of the
whole cluster, as should be expected if this is the dominant cluster galaxy.
If we consider the morphological types, the mean velocities and corrected
velocity dispersions are:
,
for E + S0 galaxies (48 objects) and
,
for
S + I galaxies (22 objects). As what is observed in most clusters, the
velocity dispersion of the late-type population tends to be larger than
that of the early-type population (Sodré et al. 1989; Stein 1997; Carlberg
et al. 1997; Adami et al. 1998). Figure 6 presents the radial velocity
distribution of the cluster galaxies as well as a Gaussian curve with the
same mean velocity and dispersion observed for these galaxies between 12 000
and
.
Note that the velocity distribution obtained with a larger
set of velocities does not confirm the relative depletion of galaxies with
proposed by Capelato et al. (1991).
![]() |
Figure 6:
The radial velocity distribution for the Abell 376 sample of
galaxies between 12 000 and
![]() |
Open with DEXTER |
Abell 376 is an average X-ray source, first detected by the Uhuru satellite (Kellogg et al. 1973). It was then observed by Einstein IPC, Exosat, ROSAT HRI and PSPC (part of the ROSAT all-sky survey), and ASCA. Abell 376 is included in the XBACs catalogue (X-ray Brightest Abell type Cluster, Ebeling et al. 1996).
The physical properties of the Abell 376 intracluster medium (ICM) were first
obtained with Exosat data in the [0.1-20.0] keV band (Edge &
Stewart 1991). Using the RS thermal plasma model (Raymond & Smith 1977) to
fit the Exosat X-ray spectrum, they derived a mean temperature for
the ICM,
kT = 5.1+3.2-1.9 keV (90% confidence) and an X-ray flux in
the [2.0-10.0] keV band,
erg cm-2 s-1, which corresponds to a bolometric luminosity
of
erg s-1. The
deprojection analysis of the Exosat image suggested a mild cooling
flow, with a mass flow rate of
yr-1 and a
cooling radius
kpc.
The ROSAT PSPC observation of Abell 376 was analyzed as part of the XBACs
catalogue (Ebeling et al. 1996). The temperature was fixed using the EXOSAT
value (see above) and the derived flux (assuming a fixed metallicity of
)
was
erg cm-2 s-1 in the
[2.0-10.0] keV band.
White et al. (1997) analyzed the Einstein IPC X-ray image of Abell 376
with a deprojection technique. In order to be applied, they used the ICM
temperature given by Edge & Stewart (1991) as an input to the deprojection.
The resultant temperature was thus
keV and
the bolometric luminosity,
erg s-1. They also derived a weak cooling-flow, with a mass
deposit rate of
yr-1 inside a cooling
radius of
kpc, well in agreement
with the previous result based on Exosat data.
The imaging analysis also yields the gas and dynamical masses. Within the
outer radius, where the observation was meaningful (
kpc), White et al. (1997) estimated
and
.
We have downloaded the ROSAT HRI and ASCA GIS data files. From the HEASARC Online Service, as far as we know, these are still unpublished data. With the HRI data, we created an exposure-map, vignetting corrected broad-band image following the "cookbook'' from Snowden & Kuntz (1998); this image was then adaptatively smoothed.
The ASCA GIS spectrum and image were extracted and cleaned with XSELECT 1.4b
following the "ASCA data reduction
guide''.
We have used the GIS background files provided by the ASCA guest observer
facility (GOF). For the spectroscopy, the GIS data was selected in a circle of
8.1 arcmin centered in the BCG and in the [0.2-10.5] keV energy band, then
rebinned with a minimum of 30 counts per energy bin.
![]() |
Figure 7: Left: Red DSS image overlayed with ROSAT HRI isocontours. Right: ASCA GIS broad band smoothed image superposed with HRIisocontours. |
Open with DEXTER |
![]() |
Figure 8:
ROSAT HRI isocontours (continuous lines) as in
Fig. 7
superposed to the isopleths of the projected galaxy distribution (dashed
lines, galaxies within
![]() ![]() |
Open with DEXTER |
In Fig. 7 we show both an optical DSS and ASCA X-ray image superposed with the smoothed HRI image isocontours. The X-ray centre as determined by the HRI coincides with the central elliptical galaxy of Abell 376. There is a good agreement between HRI and GIS images, although there seem to be a small offset between them. Such an offset is actually comparable to the known uncertainty of the ASCA aspect solution.
The X-ray images, ROSAT HRI and ASCA GIS, probe only the central 15 arcmin region and they show no evident sign of substructure. The X-ray
isocontours are not spherical or elliptical and are elongated roughly in a
Northeast-Southwest direction. Comparison between the X-ray emission and the
cluster galaxy distribution (Fig. 8) shows a good
qualitative agreement between both gas and galaxy spatial distribution, when
considering only objects within
.
This agreement
may extend further on, since the orientation of the external isopleths seen in
Fig. 4 are close to the inner contours. When we
take into account all APS galaxies brighter than
,
there is an
offset between X-ray emission and galaxy distribution close to the
North-South direction.
The ASCA X-ray spectrum was fitted with a single temperature, absorbed MEKAL
model (Kaastra & Mewe 1993; Liedahl et al. 1995) using XSPEC 11. We have
tried two fits fixing the hydrogen column density to the galactic value,
cm-2 (Dickey & Lockman 1990), and
allowing it to vary as a free parameter. The results are summarized in
Table 2 and shown in Fig. 9
The mean temperature obtained depends whether
is fixed or not (a
well known anti-correlation between spectral fitted temperature and
), but are within error bars. The temperatures we got are slightly cooler
than the ones obtained with ROSAT and Exosat. The same happens with
the metal abundance; when
is fixed the obtained metallicity is lower.
The derived flux and luminosity are about the same with the two fits in
Table 2: in the [2.0-10.0 keV] band we have
erg s-1 cm-2 and
erg s-1.
Using the scaling relations obtained by Xue & Wu (2000) we have
(with the LX-
relation), and
(with the TX-
relation). Here we have
adopted the mean temperature of the two fits, kT = 4.0 keV. The value of
is, however, ill constrained and depends on the few spectral
points below
1 keV and a better value for the temperature may be the
one from the first fit, kT = 4.3 keV. In this case,
.
Table 2:
Fitting results of the ASCA GIS spectrum. Errors bars are 90%
confidence level. The last column gives the
over the number of
degrees of freedom (d.o.f.).
It has been shown that it is possible to classify galaxies from their spectra, and that this spectral classification correlates well with morphological types (e.g., Folkes et al. 1996; Sodré & Cuevas 1997; Madgwick et al. 2002). Since for Abell 376 we have the morphological classification done by Dressler (1980), it is interesting to verify how his Hubble types compare with spectral types obtained from galaxy spectra.
The basis of spectral classification is that normal galaxy spectra form a well defined sequence in the spectral space spanned by the M-dimensional vectors that contain the spectra, each vector being the flux of a galaxy (or a scaled version of it) sampled at M wavelengths. We define the spectral type of a galaxy by its position along the spectral sequence, identified through a principal component analysis (PCA) of the galaxy spectra (see Sodré & Cuevas 1997 and Sodré & Stasinska 1999 for more details).
Here, the analysis has been done using 54 spectra of cluster
galaxies with high signal-to-noise ratios. Initially the spectra
were shifted to the rest frame and re-sampled in the wavelength
interval from 3784 Å to 6100 Å, in equal bins of 2 Å.
After, we removed from the analysis 8 regions of 20 Å
each centered at wavelengths that may contain emission lines,
because the inclusion of emission lines in the analysis increases
the dispersion of the spectral sequence. The spectra were then
normalized to the same mean flux (
)
and the mean spectrum was computed and removed from the galaxy
spectra. PCA was then used to obtain the principal components.
Figure 9 shows the projection of the 54 spectra on to the plane
defined by the first two principal components. The first principal
component contains 39% of the sample variance, whereas the second
component carries 7% of the variance. We identify the spectral
type ST of a galaxy with the value of its first principal
component. In Fig. 9 different symbols correspond to different
morphological types: filled circles, open squares, open triangles,
and crosses represent E, S0, S0/S, and S galaxies, respectively.
The mean values of ST for each of these 4 classes are
,
,
,
and
,
respectively. This result confirms that
the spectral sequence is closed related to the morphological
sequence, since the ranking of galaxies of different morphological
types along the first principal component is analogous to that
found in the Hubble sequence, despite the strong overlap in the
values of ST between different classes. Note that although the
values of the principal components are dependent on the data and
its pre-processing, the ranking of galaxies along the first
principal component is to a large extent independent of the
pre-processing adopted.
![]() |
Figure 9:
ASCA GIS spectrum showing the MEKAL plasma model best fit. The
fit shown here have a fixed
![]() |
Open with DEXTER |
![]() |
Figure 10: Projection of the spectra of 54 cluster galaxies on to the plane defined by the first two principal components. The spectral sequence follows the first component and a spectral type, ST, is attributed to each galaxy by its value of this component. Filled circles, open squares, open triangles, and crosses represent E, S0, S0/S, and S galaxies, respectively. |
Open with DEXTER |
We have estimated the cluster mass inside its virial radius with the virial
mass estimator (VME), which gives less biased mass estimates when the system
is not completely sampled (e.g., Aceves & Perea 1999). The virial radius of
Abell 376 was estimated through an iterative procedure with the relation
Mpc (Girardi et al. 1998), where
is the projected velocity dispersion.
Initially, we estimated the virial radius from the 62 galaxies with radial
velocities between 12 000 km s-1 and 17 000 km s-1. Then, considering
only galaxies within the virial radius,
was re-evaluated and a new
virial radius was computed. New iterations do not changed the value of the
virial radius,
Mpc (
42 arcmin), which
is larger than the 28 arcmin (
1.6 h70-1 Mpc) central region where
radial velocities were measured.
The resulting mass within 0.7 RV is
,
where the error is a
standard deviation
computed with the bootstrap method. Note that the VME assumes that the system
is virialized, but substructures or large-scale flows tend to increase the
velocity dispersion of the cluster galaxies, leading to an overestimation of
the cluster mass. In a radius of 28 arcmin, the only feature visible is the
SW substructure, that is not even detected in the velocity-space (see
Sect. 4),
In this paper, we have reported new velocity measurements on the Abell 376
cluster of galaxies and an X-ray analysis of previously unpublished ASCA data.
This allowed us to define almost complete samples of galaxies in the central
part of the cluster, and to discuss the dynamical status of this cluster. The
study of the galaxy projected positions reveals a regular distribution,
centered on an E/D dominant galaxy. An analysis with the adaptive kernel
density map indicates the presence of a substructure SW of the cluster main
galaxy concentration which is not confirmed by the velocity field, implying
that it may be a fortuitous condensation of galaxies bound to Abell 376. Our
dynamical analysis give for Abell 376 a virial mass about
within 70% of the virial radius of the cluster.
The velocity dispersion derived from the X-ray analysis is in agreement with
the galaxy velocity distribution. We conclude that Abell 376 is a
bona fide relaxed cluster inside
900 h70-1 kpc or
40% of the
virial radius, fairly relaxed inside 0.7 RV with only a minor
substructure, and clumpier as we reach the virial radius, where possible
substructures are detected in the galaxy distribution.
Acknowledgements
We thank the OHP and Pic du Midi staff for their assistance during the observations. HVC, LSJ and GBLN are grateful for the financial support provided by FAPESP and CNPq. DP acknowledges the France-Brazil PICS-1080 and IAG/USP for its hospitality.